It has been shown in vitro that human oral and intestinal bacteria are able to methylate inorganic mercury (Hg) to methylmercury (MeHg). In vivo methylation has been reported for guinea pigs, and several studies have found higher levels of organic Hg in blood of workers exposed to inorganic Hg. In search for indications of in vivo methylation of Hg, we determined total Hg in urine (U), plasma (P), and erythrocytes (Ery), and MeHg in erythrocytes, in 22 chloralkali workers exposed to inorganic Hg, and in 22 age-matched referents. The subjects in both groups ate fish not more than once a week. MeHg was determined using ion exchange separation and cold vapor atomic absorption spectrometry. The mean concentrations of total mercury in chloralkali workers and referents were 26 vs 2.3 micrograms/g creatinine (U-Hg), 9.7 vs 1.8 ng/g (P-Hg), and 11.4 vs 5.6 ng/g (Ery-Hg). The mean Ery-MeHg concentrations were 3.0 vs 3.7 ng/g, and there were no correlations between Ery-MeHg on the one hand and U-Hg or P-Hg on the other. Our results indicate no significant in vivo methylation of inorganic Hg in man. Contradictory results from previous studies may be due either to bias owing to fish consumption or to analytical shortcomings.